Graph-based Embedding Smoothing for Sequential Recommendation
ABSTARCT :
In real-world scenarios, a user's interactions with items could be formalized as a behavior sequence, indicating his/her dynamic and evolutionary preferences. To this end, a series of recent efforts in recommender systems aim at improving recommendation performance by considering the sequential information. However, impacts of sequential behavior on future interactions may vary greatly in different scenarios. Additionally, semantic item relations underlying item attributes have not been well exploited in sequential recommendation models, which could be crucial for measuring item similarities in recommendation. To deal with the above problems, this paper provides a general embedding smoothing framework for sequential recommendation models. Specifically, we first construct a hybrid item graph by fusing sequential item relations derived from user-item interactions with semantic item relations built upon item attributes. Second, we perform graph convolutions on the hybrid item graph to generate smoothed item embedding. Finally, we equip sequential recommendation models with the smoothed item representations to enhance their performances. Experimental results demonstrate that with our embedding smoothing framework, the state-of-the-art sequential recommendation model, SASRec, achieves superior performance to most baseline methods on three real-world datasets. Moreover, the results show that most mainstream sequential recommendation models could benefit from our framework.
EXISTING SYSTEM :
? We propose a systematic classification schema to organize the existing GNN-based recommendation models. We first categorize the existing works into general recommendation and sequential recommendation based on the task they deal with.
? Most of the existing works in GNN-based recommendation adopt ConvGNN to simulate the propagation process.
? We further categorize them into three subcategories based on the types of information used. Without side information, existing models consider the useritem relationships as a user-item bipartite graph.
? Most of the existing works capture users’ dynamic preferences only based on the sequences.
DISADVANTAGE :
? Modeling users’ preference from his historical sequences is one of the core problem of sequential recommendation.
? With dramatic growth of the amount of information on the Internet, recommender systems have been applied to help users alleviate the problem of information overload in online services, such as e-commerce, search engines, and social media.
? SASRec applies self-attention mechanisms to sequential recommendation problems to explicitly model the relationship between items.
? To settle this problem, we build a message propagation mechanism from user to item, considering the order information.
PROPOSED SYSTEM :
• In order to adapt to different scenarios, various strategies are proposed to better integrate the two representations, such as GRU mechanism, concatenation with nonlinear transformation and sum operation.
• Due to the specific characteristic of different types of data in recommendation, a variety of models have been proposed to effectively learn their pattern for better recommendation results, which is a big challenge for the model design.
• With the emergence of online social networks, social recommender systems have been proposed to utilize each user’s local neighbors’ preferences to enhance user modeling.
ADVANTAGE :
? DGSR-GCN achieves the poorest performance of all variants. It is probably because the GCN net is a linear approximation of localized graph convolution, assuming that all interactions contribute equally.
? Compared with the excellent performance in the session-based recommendation scenario, the performance of SR-GNN is flat in the sequential recommendation and especially poorly in ML-1M data.
? Increasing the layer of DGSR is capable of promoting the performance substantially. It demonstrates that exploiting high-order user sequences information explicitly can effectively improve recommendation performance.
? With the increase of ??, the model performance tends to be stable because of the limitation of the number of DGAN layers.
|